Keynote & Invited Speakers
Keynote Speakers
Keynote Speaker I
Prof. Jie Yang, Shanghai Jiao Tong University, China
Bio: Jie Yang received a bachelor’s degree in Automatic Control in Shanghai Jiao Tong University (SJTU), where a master’s degree in Pattern Recognition & Intelligent System was achieved three years later. In 1994, he received Ph.D. at Department of Computer Science, University of Hamburg, Germany. Now he is the Professor and Director of Institute of Image Processing and Pattern recognition in Shanghai Jiao Tong University. He is the principal investigator of more than 30 national and ministry scientific research projects in image processing, pattern recognition, data mining, and artificial intelligence. He has published six books，more than five hundreds of articles in national or international academic journals and conferences. Google citation 24964，H-index 81. Up to now, he has supervised 6 postdoctoral, 48 doctors and 70 masters, awarded six research achievement prizes from ministry of Education, China and Shanghai municipality. He has owned 48 patents. Three Ph.D. dissertation he supervised was evaluated as “National Best Ph.D. Dissertation” in 2009, in 2017, in 2019. He has been chairman and keynote speaker of more than 10 international conferences.
Speech Title: Researches on the Defenses
and Out-of-Distribution Detection in Trustworthy Deep Learning
Speech Abstract: The rapid
advancement of deep learning has had a transformative effect on the
development of technology and society across a multitude of sectors. In
safety-critical contexts, the potential for neural network models to produce
unreliable outputs in response to “malicious” or “unanticipated” inputs
poses a severe risk. This talk delves into the output reliability from
neural network models within the domain of trustworthy deep learning. 1)
Inputs that involve pixel perturbations, exemplified by adversarial
examples,w.r.t the task of adversarial and certified robustness; 2) Inputs
that represent distribution shifts, exemplified by Out-of-Distribution (OoD)
data, w.r.t the task of out-of-distribution detection. We introduce a novel
strategy of model augmentation, adopt a multi-head neural network structure,
and pose diversity constraints related to adversarial robustness into the
model parameters.We adopt a multi-head neural network structure, use the
ensemble of multiple heads in place of the ensemble of multiple neural
networks,which significantly reduces the computational load in both training
and certification phases. We propose that the non-linearity in InD and OoD
data hinders PCA from learning a subspace that fully embodies their
diversities. We propose a mode ensemble method that not only enhances
detection performance but also significantly reduces the performance
variance among independent modes.We propose performing linear dimension
reduction on the gradient using a designated subspace that comprises
principal components.
Papers about above topics have been published in TPAMI, IJCV, PR, NIPS in
recent years.
Keynote Speaker II
Prof. Hiroshi Fujita, Gifu University, Japan
Bio: Prof.
Hiroshi Fujita received his Ph.D. from Nagoya University, Japan, in 1983.
From 1983 to 1986, he was a visiting researcher at the Kurt Rossmann
Laboratories for Radiologic Image Research at the University of Chicago,
USA. He became an Associate Professor in 1991 and a Professor in 1995 in the
Faculty of Engineering at Gifu University. In 2002, he was appointed
Professor and Chair of Intelligent Image Information at the Graduate School
of Medicine, Gifu University. He currently serves as a Research Professor
and holds the title of Professor Emeritus at Gifu University. He is an
Honorary President of the Society for Medical Image Information (MII), a
Fellow of the Institute of Electronics, Information and Communication
Engineers (IEICE), and an Honorary Member of the Japan Society for Medical
Imaging Technology. He has served as a conference chair, invited speaker,
and keynote speaker at numerous national and international scientific
conferences. His research interests include computer-aided diagnosis (CAD),
AI in medical imaging, medical image analysis, processing, and evaluation,
as well as emerging technologies such as generative AI, vision-language
models (VLMs), and AI agents for medical applications. He has received
numerous awards and honors, including the Medical Imaging Information
Society Award and multiple awards from the Radiological Society of North
America (RSNA) and the International Society for Optics and Photonics
(SPIE). He has authored or co-authored over 1,000 publications, including
scientific journal articles, international conference papers, book chapters,
and other scholarly publications. He has been repeatedly recognized among
the world's top 2% most-cited scientists in both the "Career-long Impact"
and "Single-year Impact" categories of the Stanford–Elsevier ranking,
including in the latest 2026 edition.
Speech Title: The Evolution of AI in
Medical Imaging: From Deep Learning to Generative AI, AI Agents, and Beyond.
Invited Speakers
Invited Speaker I
Associate Professor Nian Wang, Advanced Imaging Research Center, UT Southwestern Medical Center, Dallas, Texas, USA
Bio: Dr. Nian Wang is an Associate Professor in the Advanced Imaging Research Center at UT Southwestern Medical Center. My expertise is in developing novel quantitative MRI techniques and analysis methods on the CNS system and musculoskeletal system. His lab has extended the research on the technical development and methodology innovation in medical imaging and Artificial Intelligence (AI) and their translation to MRI, x-ray, and optical imaging. He has published more than 70 papers in reputed journals and been invited to devliver talks in many international conferences.
Speech Title: High-resolution MRI Predicted Whole Mouse Brain Cell Type Atlas using Multi-modal Fusion Network (MFNet)
Speech Abstract: Mapping brain cell types with high spatial precision is fundamental for understanding brain structure and function. Traditional cell atlases, such as the Allen Brain Cell Atlas, rely on labor-intensive techniques like single-cell sequencing and spatial transcriptomics, often at limited spatial resolution. In this study, we present a deep learning framework—Multi-modal Fusion Network (MFNet)—that predicts neuronal cell types across the whole mouse brain by integrating high-resolution diffusion MRI (dMRI) and multi-modal datasets from the BICCN mouse brain atlas. Using the Allen Brain Cell Atlas as ground truth, we trained MFNet to predict dominant neuronal cell neighborhoods and classes, taking advantage of regional specificity and hierarchical cell-type structures. High-resolution (45 µm isotropic) ex vivo dMRI was acquired and processed with multiple diffusion models (DTI, DKI, NODDI, SANDI), then registered to the Allen Mouse Brain Common Coordinate Framework (CCFv3), along with spatial transcriptomics and epigenomics data. MFNet produces a full-scale cell type atlas at 10 µm isotropic resolution, significantly enhancing the spatial detail of the original atlas. Additionally, we examined correlations between dMRI features and gene expression of marker genes. This study offers a novel, efficient method for generating high-resolution brain cell atlases and demonstrates the potential of advanced imaging combined with deep learning to decode brain cellular architecture and its relevance to development and disease.
Invited Speaker II
Associate Professor Yanlei Gu, Hiroshima University, Japan
Bio: Dr. Yanlei Gu received his Ph.D. from Nagoya
University. He subsequently worked as a researcher at the University of
Tokyo and was awarded a Japan Society for the Promotion of Science (JSPS)
Postdoctoral Fellowship for Overseas Researchers. In 2019, he joined
Ritsumeikan University as a Lecturer, and since 2024, he has been an
Associate Professor at Hiroshima University. He also serves as a Visiting
Researcher at the University of Tokyo.
His research interests include Autonomous Driving, Intelligent
Transportation Systems (ITS), Machine Learning, and Computer Vision. He has
published more than 120 peer-reviewed journal and international conference
papers. His research has received several awards, including Best Paper
Awards and other honors from IEEE and international conferences.
He is a member of the IEEE Intelligent Transportation Systems Society and
the IEEE Consumer Technology Society. He has served on the organizing
committees of IEEE ICVES 2015 and IEEE ITSC 2017, as well as on the
technical program committees of ICIPRob 2021, 2022, 2024, and 2026.
Speech Title: Vision-Based Pedestrian
and Cyclist Intention Recognition for Autonomous Driving
Speech Abstract:
Understanding the behavior and intentions of pedestrians and cyclists
is essential for safe autonomous driving in complex traffic environments. In
particular, accurately recognizing their crossing intentions enables
autonomous vehicles to anticipate potential risks and make appropriate
driving decisions. Recent advances in computer vision and deep learning have
enabled the analysis of road users' appearance, body posture, movement
patterns, and surrounding traffic conditions using visual information.
However, intention recognition remains challenging due to variations in
human behavior, occlusions, environmental conditions, and the uncertainty of
human decision-making. This talk presents our research on vision-based
intention recognition for pedestrians and cyclists, with a particular focus
on predicting their crossing intentions. It discusses the use of visual
features, temporal behavior analysis, and contextual information to
understand and predict their actions. Key challenges, including early
intention prediction, behavior interpretation in different traffic
scenarios, and the integration of contextual information, are also examined.
Finally, future research directions toward more reliable and context-aware
intention recognition for autonomous driving are discussed.
Invited Speaker III
Associate Professor Kazuya Ueki, Meisei University, Japan
Bio: He received his B.S. degree in Information Engineering in 1997 and his M.S. degree in Computer and Mathematical Sciences in 1999, both from Tohoku University, Sendai, Japan. In 1999, he joined NEC Soft, Ltd., Tokyo, Japan, where he was mainly engaged in research on face recognition. He received his Ph.D. degree from the Graduate School of Science and Engineering, Waseda University, Tokyo, Japan, in 2007. From 2013 to 2017, he served as an Assistant Professor at Waseda University. He is currently an Associate Professor in the School of Information Science, Meisei University. His research interests include information retrieval, video anomaly detection, pattern recognition, and machine learning. He is involved in the video retrieval evaluation benchmark (TRECVID) sponsored by the National Institute of Standards and Technology (NIST), contributing to the development of video retrieval technology. His submitted systems achieved the highest performance in the TRECVID AVS task in 2016, 2017, 2022, and 2025.
Speech Title: Lessons from the Front Lines of Ad-hoc Video Search
Speech Abstract:The Ad-hoc Video Search (AVS) task in the TRECVID benchmark has been a demanding testbed for video retrieval, where systems must find relevant shots from free-form text queries across large video collections. Over the years, the field has moved from predefined visual concepts to CLIP-style text–video embeddings, and now to vision-language models (VLMs) that interpret user intent and reason about visual content. This talk shares what we learned while competing through these shifts, focusing on three approaches: query expansion using large language models and image generation, caption-driven domain adaptation without manual annotation, and VLM-based semantic verification that re-ranks results through visual question answering. Drawing on experimental findings and hands-on experience, the talk highlights what made a difference in practice and discusses the challenges that remain for future video retrieval.
Invited Speaker IV
Associate Professor Zichi Wang, Shanghai University, China
Bio: Zichi Wang received the BS degree in
electronics and information engineering from Shanghai University, China, in
2014, and received the MS degree in signal and information processing in
2017, the PhD degree in information and communication engineering from the
same university in 2020. He is currently with the School of Communication
and Information Engineering, Shanghai University, Shanghai, as an Associate
professor. His research interests include steganography, steganalysis, and
artificial intelligence security. He has published over 100 papers in these
areas. He has served as an invited speaker for several international
conferences, including ACMLC 2026 and AIintelli Vision 2027, served as a
program chair of international conference AAIP 2027, served as chair of
special session for several international conferences, including AAIP 2025,
AAIP 2026, and IEEE PRMVAI 2026 and ACMLC 2026, and served as a TPC member
for several international conferences, including ICECI 2022, AAIP 2025, CCIE
2026, IEEE PRMVAI 2026, ACMLC 2026 and ICITAC 2027.
Speech Title: Steganography in Neural Networks.
Speech Abstract: Steganography aims to achieve covert communication
by hiding secret data into a normal cover and transmitting it over a public
channel, which is significant for national information security. Traditional
steganography transmits secret data through multimedia such as image, audio
and video. Neural network model is a new type of data which grows rapidly in
recent years and is widely used. Steganography for neural network models is
an emerging research field that needs to be developed. Compared with
multimedia, neural network model has complex structure, wide variety and
large number of parameters. For this reason, traditional steganographic
methods cannot be used for neural network model directly. This speech will
discuss the secure steganography for neural network models.
Invited Speaker V
Dr. Chiharu Kai, Fujita Health University, Japan
Bio: She received her M.S. degree in Health Sciences from Kumamoto University in 2019. She joined Konica Minolta, Inc. in 2019, where she was engaged in research on medical image processing. She served as an Assistant Professor at Niigata University of Health and Welfare from 2022 to 2025, where she also received Ph.D. degree in 2025. She is currently an Assistant Professor in the School of Medical Sciences at Fujita Health University. Her research interests include medical image analysis, artificial intelligence (AI), and computer-aided diagnosis (CAD). Her research focuses on developing predictive AI systems for early breast cancer detection using mammograms.